Specifying low drying shrinkage crack control

Drying shrinkage has been ineffectively used by some design engineers as a means of controlling cracking in slab construction. The AS1012.13 drying shrinkage test measures the shortening of small concrete prisms due to drying from a water-saturated condition at the age of 7-days for various periods of drying with a standard drying period of 56 days. It therefore measures a later-age shrinkage. There are signicant differences between the mechanisms and magnitude of early shrinkage and the standarddrying shrinkage. It is the early-age shrinkage that inuences cracking. See Early-age Shrinkage of Concrete.
1000 900 800 700 600 SHRINKAGE STRAIN () 500 400 300 200 100 0
day 8450 day 27 1day day 8450 day 27 1day day 17

(Kraai, 1982, Appleyard, 2010). Early shortening is largely due to early shrinkage and thermal contraction which can only be controlled by good mix design, proper compaction and nishing and curing. Restraint can be minimised through good sub-base grading and nishing; placement of membrane; installation of bond separation material, installation of dowel or key joints, timing of saw cut joints, andcuring. See Avoiding Early Cracking. In prestressed concrete design and construction, drying shrinkage limits are not as important in design as minimising restraints due to the shortening of post-tensioning slabs by vertical supporting elements. Such restraints are best controlled by positioning of the vertical elements, placement of permanent and temporary movement joints, and isolation of shear walls (Cross, 2010). In watertight structures, early slab stressing, proportioning of the concrete mix and protection of slabs during its early life are important to minimise early shrinkage. In the design of post-tensioned structural elements, the impact of drying shrinkage on longterm deection, prestressing losses and amount of shrinkage and temperature reinforcements are of interest. The reduction of the AS1012.13drying shrinkage limit from 800 to 500 results in small 3% and 6% reductions in prestressing loss and long-term deection respectively, and no impact on the average intensity of effective prestress in concrete (Cross, 2010). The specication by engineers of unrealistically low standard drying shrinkage is potentially difcult to achieve as well as adversely affecting the workability of the concrete (Butcher, 2010). A careful assessment of the appropriateness of having a lower level of drying shrinkage than is commonly found in an area may avoid unnecessary costs. If lower drying shrinkage limits than would otherwise be achieved are specied, the cost of concrete will in most cases increase due to the need to use more-expensive aggregates or placing methods. This cost increase will vary from one area to another because of the different properties of the local naturally-occurring aggregate deposits. References Appleyard L Impact on Structural Design, Presentation at the Concrete Institute of Australia National seminar series on Shrinkage, Cracking and Restraint, What comes rst? March 2010. Butcher R Impact on Concrete Supply, Presentation at the Concrete Institute of Australia National seminar series on Shrinkage, Cracking and Restraint, What comes rst? March 2010. Cross E Impact on Post-tensioning, Presentations at the Concrete Institute of Australia National seminar series on Shrinkage, Cracking and Restraint, What comes rst? March 2010. Dux P, OMoore L and Dao V Crack Control: Are we getting it right, Proc Shrinkage, Cracking and Restraint: What Comes First?, Concrete Institute of Australia, March 2010. Kraai P P Concrete Drying Shrinkage: Facts or fallacies, Design for Creep and Shrinkage in Concrete Structures. American Concrete Institute SP-76, 1982, pp 2551.

day 863 day 8450 day 27 1day

day 863

Centre of test panel

Restraint-free box

AS 1012.13

Queensland University insitu and standard shrinkage results Research conducted at the University of Queensland showed that a concrete with a standard drying shrinkage of 650, used in a slab construction, exhibited a 15-month total shrinkage strain at the centre of a test panel of around 225, out of which 125 or 56% occurred in the rst week. Cracking due to plastic shrinkage and plastic settlement usually occurs within the rst day from casting, whereas cracking due to early thermal contraction mainly occurs within the rst few weeks from casting. In a companion restraint-free box, the same concrete exhibited a 15-month shrinkage strain of around 370 , out of which 295 or 80% occurred in the rst week before the commencement of strain measurement in the restraint-free standard drying shrinkage test. These clearly demonstrate the standard drying shrinkage to be signicantly higher than actual later-age drying shrinkage (day 8450) and bears no relationship to the early-age shrinkage considered critical to crack control. The results conrmed that the specication of lower standard drying shrinkage strains alone will not control or prevent the development of early-age tensile stresses that lead to earlyage cracking, and that adequate protection and curing are the only means of controlling the development of restrained tensile stresses during the rst few hours after casting, and hence reduce the risk of unplanned cracking (Dux et al, 2010). In slab construction, it has long been recognised that good construction practices that reduce early shrinkage and minimise restraint are essential requirements in achieving the construction of relatively crack-free slabs